Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    Considerations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    始まった
    0.46
    YNAMIC
    0.40
     Cost
    0.40
     persisted
    0.40
     hypotheses
    0.40
    oglob
    0.39
     Despite
    0.39
     begann
    0.39
    onan
    0.38
     despite
    0.38
    POSITIVE LOGITS
    যোগ
    0.45
     తెర
    0.44
    ệ
    0.40
    废
    0.39
     aimer
    0.38
    廢
    0.38
     priem
    0.37
    roffenen
    0.37
    قد
    0.36
     Surge
    0.36
    Act Density 0.001%

    No Known Activations